Regulatory Capital, RWA, Leverage, and Liquidity Requirements Under Stress

Chapter 6

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Under Stress Chapter Outline 6.1 Introduction 6.2 Regulatory Capital 6.2.1 How to Compute the Regulatory Capital 6.2.2 Bank Alpha’s Stress Testing Regulatory Capital 6.3 Risk-Weighted Assets and Capital Ratios 6.3.1 Bank Alpha’s Risk-Weighted Assets (From a Silo Perspective) 6.3.2 Risk-Weighted Asset Aggregation 6.3.3 Bank Alpha’s Stress Testing Capital Ratios

196 198 199

203 208

208 211 213

6.4 Leverage and Liquidity Ratios 216 6.4.1 Leverage Ratio 216 6.4.2 Bank Alpha’s Stress Testing Leverage 217 6.4.3 Liquidity Coverage Ratio 219 6.4.4 Bank Alpha’s Stress Testing Liquidity Coverage Ratio 220 6.4.5 Net Stable Funding Ratio 222 6.4.6 Bank Alpha’s Stress Testing Net Stable Funding Ratio 224 6.5 Summary 227 Suggestions for Further Reading 232 Exercises 232 References 234

Growing attention has been devoted in recent years to banks’ capital adequacy, liquidity willingness, and balance sheet integrity. Banks are required both to improve the quality of their resources and to enforce their capital ratios. This chapter brings together all components examined in the previous chapters to compute synthetic measures of solvency. Additionally, it builds a bridge toward the risk integration and reverse stress testing processes described in Chapters 7 and 8. Regulators use capital ratios to assess bank resilience. In this regard, it is useful to investigate how regulatory capital is defined within the Basel III Accord frame. Starting from an accounting representation, we apply a series of adjustments to compute regulatory capital layers. The differentiation among these tiers is based on the degree of dilution of own funds compared with equity (i.e., common shares, retained earnings, and so on). Stress Testing and Risk Integration in Banks. http://dx.doi.org/10.1016/B978-0-12-803590-0.00006-0 © 2017 Elsevier Inc. All rights reserved.

195

196 Stress Testing and Risk Integration in Banks

Market, credit, and operational RWAs are studied starting from a silo standpoint. Stressed risk parameters, balance sheet projections, and profit and loss estimates are critical gears of this engine. On this subject, after a silo RWA computation, an additional step is required to align with regulatory constraints on an aggregated basis. All in all, the analysis of capital ratios finalizes the end-to-end stress testing process examined through the illustrative example of Bank Alpha. The recent financial crisis led regulators to introduce nonrisk-weighted measures to assess a bank’s capability to face adverse conditions. The last part of the chapter shows how to monitor the balance between resources and investments by means of the leverage ratio. Furthermore, liquidity is studied through the liquidity coverage ratio (LCR) and the net stable funding ratio (NSFR).

6.1 INTRODUCTION One may be dragged into the details of a specific topic within a wide stress testing exercise and miss the end-to-end picture behind it. The goal of this chapter is to wrap up the knowledge gained in the previous chapters and present stress testing core results. Chapters 2–5 outlined a series of methods, processes, and instruments used within a stress testing exercise. Market, credit, and operational risks polarized the attention. Nonetheless, a wider perspective was followed by our considering a broader range of risks beyond pillar 1 (e.g., interest rate and liquidity). It is now time to show how to use these tools to assess the overall bank resilience against adverse conditions. A regulatory perspective characterizes this chapter. In contrast, Chapters 7 and 8 will relax these restrictions by applying a wider managerial view. One of the main ideas underlying the Basel II Accord is to ensure banks have enough capital to face unexpected losses. For this purpose, the following threshold was introduced as a trigger for capital enforcement: Regulatory capital ≥ 8%(RWAmarket + RWAcredit + RWAoperational ),

(6.1)

where RWAmarket + RWAcredit + RWAoperational is the sum of market, credit, and operational RWAs. One should question why we should use RWAs instead of assets (as is common practice in finance) in Eq. (6.1). The main reason is to take into account the impact of risks. In other words, assuming that a risky asset has a 100% weight, a $100 investment on such an asset involves a minimum regulatory capital of $8. In contrast, a risk-free asset having a 0% weight does not need any capital exceeding the expected losses requirement (already included within own funds by means of provisions). Fig. 6.1 helps us understand the relationship between assets, liabilities, and regulatory capital. Assets are classified according to their risk profile to compute the RWAs. The latter may be less than or greater than the total asset value (for

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 197

Liabilities Assets RWA Tier 2 components Equity

Additional tier 1 CET1

Regulatory capital CET1

Tier 1

Tier 2

FIG. 6.1 Regulatory framework at a glance. RWAs, Risk-weighted assets.

this reason the RWA box has some thin layers on top). In contrast, regulatory capital components are ranked in line with their capability to absorb losses. Therefore common equity tier 1 (CET1) is made up of instruments such as common shares and retained earnings. Debt-like instruments with some degree of subordination are included as additional tier 1 and tier 2 components. In line with this representation, a risk-based capital ratio is computed as follows Capital ratio =

Regulatory capital , RWAs

(6.2)

where higher ratios ensure a more comprehensive risk coverage. The connection between Eqs. (6.1) and (6.2) is evident. The 8% rule applies to a total capital ratio definition as described in Section 6.3.2. The crisis experienced in 2007–09 led a regulatory enhancement. The vast bulk of Basel II remaining in operation, changes due to Basel III (BIS, 2011) focused on few specific areas. Firstly, the credit risk was explicitly included within the market risk area (e.g., counterpart credit risk, credit value adjustments). Then improvements on capital standards as well as the introduction of leverage and liquidity ratios completed the renewed frame. As part of the scheme described by Eqs. (6.1) and (6.2), Basel III revises the definition of capital and specifies new minimum capital requirements in line with Table 6.1. By Jan. 2019, the total minimum total capital requirement will increase from 8% to 10.50%. This is due to a 2.50% capital conservation buffer to absorb losses during periods of financial and economic stress. Additional buffers may be required for stress testing purposes as described in Chapter 1. As a bank’s capital falls into the buffer range and approaches the minimum requirement, the bank would be subject to increasing restrictions on earnings distribution. Basel III also establishes a countercyclical capital buffer and another cushion needs to be considered for systemically important financial institutions. Currently there is a lot of uncertainty regarding what will be classified as a systemically important financial institution and what the consequences will be. The countercyclical capital buffer is structured as an addon to the capital conservation buffer. It is meant to counterbalance procyclical bank lending behavior. This is achieved by the linking of the height of this buffer to the economic cycle. If there are signs of excessive credit growth, the buffer

198 Stress Testing and Risk Integration in Banks

TABLE 6.1 Basel Committee on Banking Supervision Road Map for Minimum Capital Requirements (the Dates Refer to Jan. 1) 2016

2017

2018

2019

Minimum common equity capital ratio (%)

4.50

4.50

4.50

4.50

Capital conservation buffer (%)

0.625

1.25

1.875

2.50

Minimum common equity plus capital conservation buffer (%)

5.125

5.75

6.375

7.00

Minimum tier 1 Capital (%)

6.00

6.00

6.00

6.00

Minimum total capital (%)

8.00

8.00

8.00

8.00

Minimum total capital plus conservation buffer (%)

8.625

9.25

9.875

10.50

can be applied at the discretion of the national regulatory authority. The buffer ranges from 0% to 2.5% and consists of either tier 1 common equity or other fully loss absorbing capital instruments. Section 6.2 focuses on the numerator of Eq. (6.2), while Section 6.3 highlights how to compute the denominator. In this regard, market, credit, and operational risk silos are aggregated to compute the overall RWAs. Furthermore, as a response to 2007–09 crisis, Basel III introduced nonriskbased indicators to monitor bank solvency. On this subject, Section 6.4 describes the leverage ratio as an indicator of a structural balance between own resources and assets. Following a liquidity view, a distinction is made between coverage and the NSFR: the former attempts to capture the short-term mismatch between cash outflows and inflows, whereas the latter focuses on a longer-term liquidity counterbalance. Unlike in the previous chapters, a process-oriented approach qualifies the following sections. The focus is on regulatory rules to compute the synthetic indices of endurance described above. In this regard, links between regulatory requirements and accounting rules are crucial for a comprehensive understanding of the entire method.

6.2 REGULATORY CAPITAL Basel I (BIS, 1988) linked the minimum capital standards to credit risk. An amendment is extended to market risk the capital requirement computation. Then, Basel II (BIS, 2006) intended to enhance international capital standards by means of internal ratings-based (IRB) methods to assess the credit risk. Operational risk was also included as an additional component of the pillar 1 risk framework. In line with the above, Basel III strengthened the rules

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 199

on capital (and introduced nonrisk-based indices based on structural leverage and liquidity). The following sections describe how to compute the regulatory capital starting from a financial reporting (accounting) representation. A stress testing perspective is followed and illustrated through Bank Alpha.

6.2.1 How to Compute the Regulatory Capital Regulatory capital rules do not necessarily align with accounting standards. Hence an adjustment is required to extrapolate regulatory capital figures from accounting reports. As a starting point, one may think of Bank Alpha’s equity introduced in Chapter 3. In this case, shareholder equity is made up of common shares, preferred shares, and retained earnings. Some additional components may be considered as part of the regulatory capital by our distinguishing among CET1, additional tier 1 components, and tier 2 capital as listed below: ●

●

●

CET1. Common equity tier 1 consists of a combination of shares and retained earnings. It is the primary and most restrictive form of regulatory capital. To qualify as tier 1, capital has to be subordinated, perpetual in nature, loss bearing, and fully paid up with no funding having come from the bank. On top of common equity, all major adjustments are applied as described below. Additional tier 1 capital. This additional layer of the tier 1 capital consists of instruments paying discretionary dividends having neither a maturity date nor an incentive to redeem them. Innovative hybrid capital instruments are phased out because of their fixed distribution percentage, nonloss absorption capabilities, and incentive to redeem them through features such as step-up clauses. Tier 2 capital. Tier 2 capital contains instruments that are capable of bearing a loss, not only in the case of default but also in the event that a bank is unable to support itself in the private market. Their contractual structure needs to allow banks to write them down or convert them into shares.

Table 6.2 outlines some of the differences between Basel II and Basel III components. This is an illustrative and nonexhaustive exemplification. Misalignments may arise between accounting and regulatory capital requirements. On this subject, one needs to apply regulatory adjustments. As an example, goodwill and intangibles are usually recognized under the current accounting standards. In contrast, these elements need to be excluded from the regulatory capital. Likewise, the (negative) difference between accounting provisions and the IRB expected loss (i.e., estimated on internal rating parameters) is deducted from the regulatory capital. Table 6.3 shows some illustrative examples of adjustments. A two-step process applies to compute capital requirements. Firstly, accounting components are qualified in terms of regulatory categories (i.e., CET1, additional tier 1 capital, and tier 2 capital). Secondly, deductions are

200 Stress Testing and Risk Integration in Banks

TABLE 6.2 Comparison Between Basel II and Basel III Regulatory Capital Treatment of Some (Major) Capital Components Capital Category

Basel II

Basel III

Common shares

CET1

CET1

Retained earnings

CET1

CET1

Innovative capital instruments

Additional tier 1 capital

Excluded and grandfathered

Noninnovative capital instruments

Included in additional tier 1 capital, subject to conditions

Subordinated debt

Tier 2 capital

Included in tier 2 capital, subject to conditions

TABLE 6.3 Regulatory Capital Adjustments (Illustrative Examples) Adjustment

Description

Goodwill and intangible assets

To be deducted from CET1

Shortfall provisions

To be deducted from CET1

Unrealized gains and losses

To be taken into account in CET1

computed. A fully comprehensive description of all cases one may encounter in practice goes beyond the scope of this section. The two phases are outlined by our focusing on the key elements of Bank Alpha’s illustrative example as listed below: 1. Identification of regulatory categories. As part of the first step, balance sheet components are examined and qualified as tier 1 or tier 2 as follows: ●

Shareholder equity: – Common shares. Common stockholder equity is included in accounting capital and consists of voting shares. This is the most desirable capital element from a supervisory perspective. Indeed, shares absorb bank losses commensurate with their accounting value. They provide a savings association with the maximum amount of financial flexibility necessary during a crisis and are qualified as CET1. – Preferred shares. Preferred shares typically entitle a holder to a fixed dividend, which is received before any common stockholders may receive dividends. As a general rule, they qualify for inclusion in tier

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 201

●

●

1 capital if losses are absorbed while the issuer operates as a going concern. Clauses, covenants, and restrictions that make these shares more debit-like may cause them not to be acceptable as tier 1 capital. – Retained earnings. Cumulated earnings due to realized profit not distributed among shareholders are usually assimilated to common shares and included within the tier 1 capital. Subordinated debts. A few characteristics are required for a debt to be qualified as subordinated and be classified as tier 2 capital. A limited repayment right usually identifies a subordinated debt in the event of default. In more detail, these financial instruments are usually limited to a nonpayment amount. The only remedy afforded to the investor is to petition for the winding up of the bank. In that case, investors are allowed to claim in the winding-up process. Additionally, the maturity needs to be longer than a certain period (e.g., 5 years) and the amortization process is required to be aligned with regulatory schemes. Noncontrolling interests. A noncontrolling interest is created when a depository institution owns a controlling interest in but not 100% of a subsidiary. The remaining interest is owned by third parties, referred to as noncontrolling shareholders. The noncontrolling interest should absorb losses in the subsidiary commensurate with the subsidiary’s capital needs. It should not represent an essentially risk-free or low-risk investment for the holders of the subsidiary capital instrument.

2. Computation of deductions. As per the regulatory capital computation process, deductions are estimated as detailed below: ●

●

Goodwill and intangible assets. Goodwill and intangibles are usually included among assets, according to accounting prescriptions. However, their intangible nature dilutes equity’s capability to absorb losses. For this reason a 100% deduction from tier 1 is required. Shortfall (provisions excess). According to the BIS (2004): in order to determine provision excesses or shortfalls, banks will need to compare the IRB measurement of expected losses EAD [exposure at default] × PD [probability of default] × LGD [loss given default] with the total amount of provisions that they have made, including both general, specific, portfolio-specific general provisions as well as eligible credit revaluation reserves discussed above. As previously mentioned, provisions or write-offs for equity exposures will not be included in this calculation. For any individual bank, this comparison will produce a “shortfall” if the expected loss exceeds the total provision, or an excess if the total provision exceeds the expected loss.

The idea behind this deduction is outlined through Fig. 6.2. As described in Chapter 4, the regulatory capital faces unexpected losses.

Frequency

202 Stress Testing and Risk Integration in Banks

(1−a) confidence level

Provisions

Expected loss (EL)

Shortfall FIG. 6.2

Losses

Unexpected loss (UL)

Shortfall mechanics.

Indeed, expected losses are supposed to be captured by provisions and deducted from equity. Therefore the difference (shortfall) between expected losses and provisions is deducted from capital. Example 6.1 details how to compute the shortfall. Example 6.1 Shortfall A bank has $8.50 billion of tier 1 capital and $5.50 billion of tier 2 capital. Its total capital accounts for $14 billion. Table 6.4 summarizes the expected loss (due to Basel IRB parameters) and provisions. A shortfall originates from the difference between these two components.

TABLE 6.4 Credit Shortfall Computed as the Difference Between Basel II Expected Loss and Accounting Provisions ($ Billions) Portfolio

Expected Loss (Basel II)

Provisions

Shortfall

Performing Nonperforming Total

10.00 25.00 35.00

9.00 22.00 31.00

1.00 3.00 4.00

According to Table 6.4, an overall $4 billion shortfall needs to be deducted from capital (for simplicity, no phasing in is considered). Therefore the tier 1 capital reduces to $4.50 billion. Likewise, the total capital drops from $14 billion to $10 billion. The reader is invited to refer to Sections 6.2.2 and 6.3.2 for a comprehensive description of the shortfall mechanics.

The next section focuses on Bank Alpha to show how to implement the above mechanism in practice.

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 203

6.2.2 Bank Alpha’s Stress Testing Regulatory Capital Bank Alpha embodies a useful example for an in-depth regulatory capital analysis. Table 6.5 summarizes Bank Alpha’s balance sheet items to be part of the regulatory capital. Data refer to t0 (i.e., stress testing starting point). The following key assumptions qualify Bank Alpha’s regulatory capital. ●

●

● ●

● ●

●

●

Subordinated debts. These instruments are assumed to be qualified as tier 2 capital. Noncontrolling interests. These interests are assumed to share tier 2 capital characteristics. Common shares. They belong to CET1. Preferred shares. All these shares are assumed to be classifiable as tier 1 capital. Retained earnings. They belong to CET1. Goodwill. Bank Alpha’s balance sheet embodies a $1.50 billion goodwill to be deducted from tier 1 capital. Shortfall performing. Following the guidelines outlined in the previous section, Bank Alpha’s expected losses on the performing portfolio are assumed to account for $0.66 billion, with a corresponding $0.60 billion collective provisions. The shortfall is $0.06 billion. Shortfall nonperforming. According to the BIS (2006), the capital requirement (K) for a defaulted exposure is equal to the greater of zero and the difference between its LGD [loss given default] and the bank’s best estimate of the expected loss. The risk-weighted asset amount for the defaulted exposure is the product of K, 12.5, and the EAD [exposure at default]. Therefore the difference between the expected loss and provisioning needs to be additionally compared against the so-called best estimate of the expected loss as detailed in Section 6.3.2.

TABLE 6.5 Bank Alpha’s Subordinated Debts, Noncontrolling Interests, and Shareholder Equity at t0 ($ Billions) t0 Subordinated debts

4.00

Noncontrolling interests

2.00

Shareholder equity

7.00 Common shares

4.00

Preferred shares

1.00

Retained earnings

2.00

t0

CET1

Additional Tier 1 Capital

Tier 2 Capital

Subordinated debts

4.00

4.00

Noncontrolling

2.00

2.00

interests Shareholder equity

7.00 Common shares

4.00

Preferred shares

1.00

Retained earnings

2.00

4.00 1.00 2.00

Deductions

Regulatory capital

Goodwill

−1.50

Shortfall

−0.06 4.44

1.00

6.00

204 Stress Testing and Risk Integration in Banks

TABLE 6.6 Bank Alpha’s Regulatory Capital From t0 to t3 ($ Billions)

t1

CET1

Additional Tier 1 Capital

Tier 2 Capital

Subordinated debts

4.00

4.00

Noncontrolling interests

2.00

2.00

Shareholder equity

6.18 Common shares

4.00

Preferred shares

1.00

Retained earnings

1.18

4.00 1.00 1.18

Deductions

Deductions Regulatory capital

Goodwill

−1.45

Shortfall

−0.16 3.57

1.00

6.00 (Continued)

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 205

TABLE 6.6 Bank Alpha’s Regulatory Capital From t0 to t3 ($ Billions)—cont’d

t2

CET1

Additional tier 1 capital

Tier 2 capital

Subordinated debts

4.00

4.00

Noncontrolling

2.00

2.00

interests Shareholder equity

6.45 Common shares

4.00

Preferred shares

1.00

Retained earnings

1.45

4.00 1.00 1.45

Deductions

Deductions Regulatory capital

Goodwill

−1.40

Shortfall

−0.26 3.79

1.00

6.00

206 Stress Testing and Risk Integration in Banks

TABLE 6.6 Bank Alpha’s Regulatory Capital From t0 to t3 ($ Billions)—cont’d

t3

CET1

Additional Tier 1 Capital

Tier 2 Capital

Subordinated debts

4.00

4.00

Noncontrolling interests

2.00

2.00

Shareholder equity

7.15 Common shares

4.00

Preferred shares

1.00

Retained earnings

2.15

4.00 1.00 2.15

Deductions

Deductions Regulatory capital

Goodwill

−1.35

Shortfall

−0.36 4.44

1.00

6.00

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 207

TABLE 6.6 Bank Alpha’s Regulatory Capital From t0 to t3 ($ Billions)—cont’d

208 Stress Testing and Risk Integration in Banks

Table 6.6 summarizes Bank Alpha’s regulatory capital starting from t0 until the end of the stress testing process (i.e., t3 ). The reader is invited to verify the alignment of these figures with the balance sheet projection described in Chapter 5. The next section is devoted to RWA analysis. Capital ratios will be estimated accordingly.

6.3 RISK-WEIGHTED ASSETS AND CAPITAL RATIOS An intricate muddle of rules presides over the RWA computation. An in-depth investigation of each norm would prevent us from showing the key features of a stress testing exercise. For this reason, the focus narrows down to Bank Alpha. This example compromises accuracy and a broad view of the process. Therefore originating from the macroeconomic scenario introduced in Chapter 2 and the balance sheet structure drawn in Chapter 3, we use Bank Alpha to explain the entire RWA estimation mechanism. In more detail, one needs to consider a stress testing transmission framework to shock risk parameters (see Chapters 3 and 4). Afterward, balance sheet and profit and loss projections are made (as detailed in Chapters 4 and 5). Finally, RWA silos are projected and an aggregation system is applied.

6.3.1 Bank Alpha’s Risk-Weighted Assets (From a Silo Perspective) Market, credit, and operational risk methods are used in combination with balance sheet and profit and loss projections to assess Bank Alpha’s stress testing RWAs. A summary of each silo RWA is shown to grasp the key features one needs to bear in mind to compute a bank’s overall RWAs under stress. The following advanced formula is used to compute the RWAs for market risk: RWAmkt = 12.5(VaRreg + SVaRreg + IRC + SSRC),

(6.3)

where the subscript mkt stands for market, VaRreg is the regulatory value at risk, SVaRreg is the stressed value at risk, IRC stands for incremental risk charge due to the counterpart credit risk. SSRC embraces the risk of loss from changes in the market value of a position that could result from factors other than market movements and includes event risk, default risk, and idiosyncratic risk. Originating from the advanced approach described in Chapter 3, Table 6.7 highlights Bank Alpha’s RWAs for market risk over the 3-year stress testing period. Stressed market risk parameters in conjunction with a small increase of the portfolio lead to a moderate RWA boost. It is worth mentioning that a dynamic balance sheet perspective implies the projection of exposures and changes in volatility due to adverse conditions over time (FRB, 2015). In contrast, a static balance sheet method anchors to the initial portfolio on which stressed parameters are applied. Table 6.7 reflects a dynamic standpoint.

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 209

TABLE 6.7 Bank Alpha’s Market Risk-Weighted Assets ($ Billions) RWAs t0

RWAs t1

RWAs t2

RWAs t3

VaR

0.30

0.42

0.41

0.40

SVaR

1.13

1.30

1.27

1.26

IRC

0.75

0.88

0.85

0.84

SSRC

0.13

0.16

0.15

0.15

Total

2.31

2.76

2.68

2.65

IRC, Incremental risk charge; RWAs, risk-weighted assets; SSRC, standard specific risk charge; SVaR, stressed value at risk; VaR, value at risk.

For the credit risk area, Table 6.8 summarizes the key features of Bank Alpha’s RWAs. The performing portfolio RWAs are computed according to the IRB advanced formula:   √ n  Φ −1 (PDi ) + ρΦ −1 (0.999) RWAcr = 12.5 · 1.06 Ai LGDi Φ − PDi adj(Mi ), √ 1−ρ i=1 (6.4) where the subscript cr stands for credit, while i identifies a debtor belonging to the (performing) portfolio,1 Φ −1 (PDi ) is the standard normal inverse cumulative distribution function, PDi is the probability of default, LGDi is the loss given default, and Ai denotes the asset exposure (i.e., exposure at default). As described in Chapter 4, correlation (ρ) and maturity adjustment (adj(M)) vary according to the type of exposure. Credit risk parameters are stressed considering the adverse scenario outlined in Chapter 2. The method explained in Chapter 4 is applied as a bridge to generate the shock. Balance sheet movements are estimated according to the schemes illustrated in Chapter 5. Table 6.8 highlights that adverse macroeconomic conditions cause a severe increase in RWAs for credit risk. The underlying risk parameters hike during the first year and partially smooth down at t2 and t3 . All in all, an approximate 66% jump is recorded at t1 compared with t0 . This spike is partially absorbed during the following years. The nonperforming portfolio is analyzed in Section 6.3.2, where a comparison between expected loss (IRB based) and the best estimate of the expected loss is conducted. This test will allow us to decide whether to record additional RWAs or reduce the regulatory capital.

1. For simplicity, the subscript s used in Chapter 4 to specify the sector to which the customer belongs is omitted.

210 Stress Testing and Risk Integration in Banks

TABLE 6.8 Bank Alpha’s Credit Risk-Weighted Assets ($ Billions) RWAs t0

RWAs t1

RWAs t2

RWAs t3

Corporate

14.69

29.38

25.85

23.27

Retail

26.90

40.35

36.32

34.50

Others

13.21

21.14

19.02

17.50

Total

54.80

90.87

81.19

75.27

RWAs, Risk-weighted assets.

As detailed in Chapter 5, Bank Alpha uses the standardized approach to estimate its operational risk. Risk weights ranging from 12% to 18% (denoted as βbl ) are applied on the gross income spread among regulatory business lines as detailed below; 8 3 t=1 max( bl=1 GIbl,t βbl , 0) , (6.5) RWAor,SA = 12.5 3 where the subscript SA stands for standardized approach and GIbl,t is the gross income of year t for business line bl. It is worth noting that t refers to the previous three years. Table 6.9 shows a feeble reduction of RWAs for operational risk during the 3-year stress testing exercise.

TABLE 6.9 Bank Alpha’s Operational Risk-Weighted Assets ($ Billions) RWAs t0

RWAs t1

RWAs t2

RWAs t3

Corporate finance

0.11

0.11

0.11

0.10

Trading and sale

0.56

0.55

0.53

0.52

Retail banking

2.40

2.34

2.28

2.22

Commercial banking

4.22

4.11

4.00

3.90

Payment and settlement

0.90

0.88

0.85

0.83

Agency services

0.28

0.27

0.27

0.26

Asset management

0.23

0.22

0.21

0.21

Retail brokerage

0.23

0.22

0.21

0.21

Total

8.93

8.69

8.46

8.24

RWAs, Risk-weighted assets.

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 211

The next section targets RWA aggregation. Moving from the regulatory capital described in Section 6.2.2 and the aggregated RWAs of Section 6.3.2, we scrutinize capital ratios in Section 6.3.3.

6.3.2 Risk-Weighted Asset Aggregation The overall capital requirements computation relies on silo RWAs. However, a more sophisticated mechanism with additional constraints holds. A regulatory floor is the first item marking the difference between a sum of silos and an aggregated measure of risk. The BIS (2015) highlights that the Basel II framework introduced a capital floor as part of the transitional arrangements for banks using the internal ratings-based (IRB) approach for credit risk and/or an advanced measurement approach (AMA) for operational risk. The objective of the floor was to ensure capital requirements did not fall below a certain percentage of banks’ capital requirements under the previous Basel I framework. In July 2009, the Committee agreed to keep in place the Basel I capital floor. Moreover, the Basel Committee on Banking Supervision views the role of a capital floor as complementing the leverage ratio introduced as part of Basel III.2 As mentioned in Section 6.2, provisions need to be examined together with RWAs and regulatory capital when IRB models are adopted. According to the BIS (2011), the capital requirement for a defaulted exposure is equal to the greater of zero and the difference between its loss given default and the bank’s best estimate of the expected loss. The RWA amount for the defaulted exposure is the product of the capital requirement, 12.5, and the exposure at default. The tenor of the above rules can be further detailed through Example 6.2. The bank introduced in Example 6.1 is further investigated to examine the regulatory floor and the impact of the best estimate of the expected loss on capital ratios.

2. The Basel framework prescribes a capital floor based on 80% of the Basel I approach for banks that apply the advanced approaches to calculate capital requirements for credit risk (IRB approach) and operational risk (advanced measurement approach). The US core banks that have exited the parallel run are required to calculate a floor based on 100% of the new US standardized approach. The US agencies have explained that for a typical US bank the US floor will be at least as conservative as the Basel I floor (BIS, 2014a). Article 500 of the EU Capital Requirements Regulation states that banks shall hold own funds that are at all times more than or equal to 80% of the total minimum amount of the own funds that the institution would be required to hold under Article 4 of Directive 93/6/EEC as that directive and Directive 2000/12/EC of the European Parliament and of the Council of Mar. 20, 2000 relating to the taking up and pursuit of the business of credit institutions stood prior to Jan. 1, 2007. The BOE (2013) refers to Article 500 of the EU Capital Requirements Regulation to apply the regulatory floor.

212 Stress Testing and Risk Integration in Banks

Example 6.2 Credit Risk and Capital Ratios: The Role of the Best Estimate of the Expected Loss The bank introduced in Example 6.1 has $8.50 billion of tier 1 capital and $5.50 billion of tier 2 capital. As in Example 6.1, the best estimate of the expected loss is assumed to be aligned with the expected loss. Therefore an overall $4 billion capital deduction is recorded as detailed in Table 6.10. As a result, the tier 1 capital reduces to $4.50 billion and the total capital becomes $10 billion.

TABLE 6.10 Shortfall When the Expected Loss Equals the Best Estimate of the Expected Loss ($ Billions) Portfolio

EL (Basel II) Provisions ELBE

Performing Nonperforming Total

10.00 25.00 35.00

9.00 22.00 31.00

RWA Nonperforming Capital Shortfall 1.00 3.00 4.00

25.00

EL, Expected loss; ELBE , best estimate of the expected loss; RWA, risk-weighted asset.

Let us additionally assume that Basel I credit and market RWAs sum up to $100 billion. By application of the 80% regulatory threshold, a floor is set up at $80 billion. Moreover, the sum of (advanced modeling) the RWAs for market, credit, and operational risk is $75 billion. In this context, the floor being higher than the sum of RWA silos, a $5 billion add-on is needed. The risk-based capital ratios when the expected loss equals the best estimate of the expected loss are computed as follows: ● Tier 1 capital ratio 4.50 80 = 5.625%. Total capital ratio 10 80 = 12.5%. Let us now modify our assumption on the best estimate of the expected loss as detailed in Table 6.11

●

TABLE 6.11 Shortfall When the Expected Loss is Greater Than the Best Estimate of the Expected Loss ($ Billions) Portfolio

EL (Basel II) Provisions ELBE

Performing Defaulted Total

10.00 25.00 35.00

9.00 22.00 31.00

RWA Nonperforming Capital Shortfall 1.00

22.00

37.50 37.50

1.00

EL, Expected loss; ELBE , best estimate of the expected loss; RWA, risk-weighted asset.

Capital deductions account for $1 billion. This causes a tier 1 capital reduction to $7.50 billion and a total capital drop to $13 billion. The difference between the expected loss and the best estimate of the expected loss (i.e., $3 billion) increases the RWAs by $3 billion ×12.5 = $37.50 billion. Therefore the sum of RWAs is $75.00 billion + $37.50 billion = $112.50 billion. In this case, the floor is crossed. No additional add-on is required. (Continued)

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 213

Example 6.2 Credit Risk and Capital Ratios: The Role of the Best Estimate of the Expected Loss—cont’d Finally, risk-based capital ratios when the expected loss is greater than the best estimate of the expected loss are listed below: 7.50 ● Tier 1 capital ratio 112.50 = 6.67%. 13.00 Total capital ratio 112.50 = 11.55%. The comparison between capital ratios in these two scenarios shows the importance of the best estimate of the expected loss in the capital assessment process.

●

The next section summarizes the overall process examined throughout the stress testing and capital requirement journey by means of Bank Alpha’s illustrative example.

6.3.3 Bank Alpha’s Stress Testing Capital Ratios Three last steps are required to compute Bank Alpha’s capital ratios. Firstly, a shortfall is assessed. Secondly, the aggregated RWAs are estimated on the basis of the Basel I floor. Finally, capital ratios are assessed by use of the results of the previous steps: ●

●

Shortfall. As a first step, Table 6.12 summarizes the key components to compute RWAs and capital shortfalls. As detailed in the previous sections, a distinction between performing and defaulted portfolios is made. Thus a comparison between the expected loss, provisions, and best estimate of the expected loss is made. Table 6.12 shows that, for the performing portfolio, the expected loss is greater than or equal to provisions during the entire stress testing exercise (i.e., t1 , t2 , t3 ). The performing EL is $0.66 billion at t0 , while provisions account for $0.60 billion. The $0.06 billion difference is the capital shortfall at t0 . This shortfall was highlighted in Table 6.6. For the defaulted portfolio, the best estimate of the expected loss is aligned with provisions. At t0 the expected loss is greater than or equal to the best estimate of the expected loss. Thus, a $2.50 billion RWA shortfall is computed (i.e., $4.60 billion − $4.40 billion = $0.20 billion × 12.5 = $2.50 billion). The same mechanism is applied during the entire stress testing exercise as detailed in Table 6.12. Provisions are aligned with those illustrated in Chapter 5 (i.e., Bank Alpha’s credit portfolio projection). Aggregated RWAs. The sum of market and credit Basel I RWAs is the starting point to estimate the aggregated RWAs (based on advanced models). A regulatory percentage is applied to Basel I RWAs to assess a floor. Table 6.13 highlights both these components. The second row outlines the floor. Therefore RWAs for market, performing credit, operational, and other risks are shown. Moreover, a nonperforming RWA is considered. The sum of all these components is compared against the floor. The difference between this latter sum and the Basel I floor, if positive, is recorded as add-

214 Stress Testing and Risk Integration in Banks

TABLE 6.12 Comparison of Bank Alpha’s Expected Loss, Provisions, and Best Estimate of the Expected Loss to Compute Risk-Weighted Asset and Capital Shortfalls Along the Stress Testing Horizon ($ Billions) Portfolio t0

t1

t2

t3

EL (Basel II)

Provisions

Performing

0.66

0.60

Nonperforming

4.60

4.40

Total

5.26

5.00

Performing

0.97

0.85

Nonperforming

6.93

6.58

Total

7.90

7.43

Performing

1.05

0.96

Nonperforming

7.87

7.74

Total

8.92

8.70

Performing

0.99

0.93

Nonperforming

8.52

8.46

Total

9.51

9.39

ELBE

RWA Capital Nonperforming Shortfall

4.40

2.50

0.06 2.50

0.06 0.12

6.58

4.38 4.38

0.12 0.09

7.74

1.63 1.63

0.09 0.06

8.46

0.75 0.75

0.06

EL, Expected loss; ELBE , best estimate of the expected loss; RWA, risk-weighted asset.

TABLE 6.13 Bank Alpha’s Overall risk-Weighted Asset Computation Along the Stress Testing Horizon ($ Billions) t0

t1

t2

t3

Basel I RWAs

95.00

101.00

103.00

104.00

Basel I floor

76.00

80.80

82.40

83.20

Market risk RWAs

2.31

2.76

2.68

2.65

Credit risk RWAs

54.80

90.87

81.19

75.27

Operational risk RWAs

8.93

8.69

8.46

8.24

Other RWAs

1.20

1.30

1.85

1.12

RWA nonperforming

2.50

4.38

1.63

0.75

69.74

108.00

RWAs (no floor) RWA add-on

6.26

Total RWAs

76.00

RWA, Risk-weighted asset.

– 108.00

95.81 –

88.03 –

95.81

88.03

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 215

TABLE 6.14 Bank Alpha’s Regulatory Capital ($ Billions), RiskWeighted Assets ($ Billions), and Capital Ratios (%) During the Stress Testing Exercise t0

t1

t2

t3

Regulatory capital ($ billions) CET1

4.44

3.57

3.79

4.44

Tier 1

5.44

4.57

4.79

5.44

11.44

10.57

10.79

11.44

76.00

108.00

95.81

88.03

Cet1 ratio

5.84

3.31

3.96

5.04

Tier 1 ratio

7.16

4.23

5.00

6.18

15.05

9.79

11.26

13.00

Total capital (tier 1 + tier 2)

RWAs Capital ratios (%)

Total capital (tier 1 + tier 2) ratio RWAs, Risk-weighted assets.

●

on RWAs. Table 6.13 pinpoints a steep increase in the total RWAs at t1 . This is mainly due to the jump of the credit risk RWAs from t0 to t1 described in Section 6.3.1. Capital ratios. Table 6.14 summarizes capital ratios during the stress testing exercise. The adverse scenario at t1 heads a harsh increase in RWAs accompanied by a dramatic reduction in capital ratios. Therefore starting from a strong capital framework at t0 , the common equity ratio falls below the 4.50% Basel III threshold. Bank Alpha also falls below the tier 1 ratio 6% threshold at t1 and t2 . This implies that Bank Alpha would probably have to consider a capitalization plan or business restructuring.

It is worth highlighting that, apart from pillar 1, other risks need to be considered to assess bank resilience against adverse conditions. In this regard, Fig. 6.3 summarizes the main components one needs to bear in mind. Common equity, tier 1 capital, and total capital ratios are the very bottom threshold. Then, bank-specific buffers are required to address pillar 2. Systemic requirements as well as stress test-specific issues (i.e., countercyclical and conservation buffers) are also included. Pillar 2 risks will be considered as part of the integration process described in Chapter 7. In what follows, the attention is devoted to leverage and liquidity ratios.

216 Stress Testing and Risk Integration in Banks

Countercyclical buffer Capital conservation buffer Systemic buffer (bank specific)

Buffers examined by stress test

Buffers set with reference to the impact of failure

Pillar 2 (bank specific)

Minimum requirements Pillar 1

FIG. 6.3

Stress testing capital buffers.

6.4 LEVERAGE AND LIQUIDITY RATIOS The leverage ratio was introduced after the 2007–09 crisis to serve as a backstop for risk-based capital ratios. The intention was to create a secondary metric that was simple and transparent to assess an appropriate balance sheet size. Moreover, the financial crisis highlighted difficulties in maintaining an adequate balance between cash outflows and inflows. Indeed, unprecedented levels of liquidity support were required from central banks to sustain the financial system. Nonetheless, even with such extensive support a number of banks failed or required resolution. These events were preceded by several years of ample liquidity in the financial system. Liquidity risk and its management did not receive an adequate level of scrutiny and priority as other risk areas. The crisis illustrated how quickly and severely liquidity risks can blow up and funding evaporate. Basel III introduced two additional regulatory standards: the LCR and the NSFR.

6.4.1 Leverage Ratio The Northern Rock case introduced in Chapter 1 highlights how an excessive leverage may be critical for bank solvency. The leverage ratio can be summarized as a measure of capital as a proportion of total adjusted assets as detailed below:

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 217

Leverage ratio =

Capital exposure . Exposure measure

(6.6)

Capital is calculated by use of the tier 1 definition described in the previous sections. In contrast, the financial accounting balance sheet is used as a starting point to measure the exposure. Specific provisions and valuation adjustments may be deducted from the exposure to which they relate. As a general principle, collateral, guarantees, and purchased credit risk mitigation may not be deducted from exposures (BIS, 2014b). A bank’s total exposure measure is the sum of the items listed below: ●

●

●

●

On-balance sheet exposures. All balance sheet assets are included according to their accounting measurement. On-balance sheet derivatives, collateral, and covenants for securities finance transactions, different from those described below, are also included. Derivative exposures. For derivatives, two types of exposures are considered. On the one hand, exposure may arise from the instrument underlying the derivative. On the other, a counterpart credit risk exposure is taken into account. All in all, derivatives are measured with use of the accounting exposure that reflects the fair value of the contract. An add-on for potential future exposure is also used to ensure a consistent conversion to a loan equivalent amount. Securities finance transactions. Secured lending and borrowing is an important source of leverage. Repurchase agreements and securities finance are included by use of the accounting measure of exposure. Regulatory netting rules are applied. Off-balance sheet items. Off-balance sheet items including commitments, letters of credit, failed transactions, and unsettled securities are subject to a uniform 100% credit conversion factor. The only exception is that any commitments that are unconditionally cancelable by the bank at any time without prior notice may have a credit conversion factor of 10%.

During the parallel run period, between 2013 and 2017, a minimum ratio of 3% is tested. The Basel Committee on Banking Supervision will investigate whether this percentage and the design of the ratio are appropriate over a full credit cycle and different types of business models. On the basis of the results for the parallel run period, there might be adjustments in the first half of 2017. The leverage ratio will become an explicit requirement as of Jan. 1, 2018. In the next section, Bank Alpha’s leverage ratio is investigated during the entire stress testing exercise.

6.4.2 Bank Alpha’s Stress Testing Leverage It is worth remarking that a few simplifications were made in Chapter 3 when Bank Alpha was introduced. One of the most important was to assume the

218 Stress Testing and Risk Integration in Banks

absence of derivatives and off-balance sheet operations. Despite the relevance of these instruments in the current economy, their exclusion did not dramatically affect the discussion. Nonetheless, one needs to bear in mind that these components may have an important role when one is assessing the leverage. In line with Bank Alpha’s balance sheet structure, a couple of additional questions arise. On the one hand, one may be puzzled by the inclusion of intangibles among assets. On the other hand, the treatment of shortfall needs to be explained more precisely in terms of its contribution to the numerator of Eq. (6.6). The BIS (2014b) addresses both these questions. Indeed, paragraph 16 states: To ensure consistency, balance sheet assets deducted from Tier 1 capital (as set out in paragraphs 66 to 89 of the Basel III framework) may be deducted from the exposure measure. Two examples follow: ●

●

Where a banking, financial or insurance entity is not included in the regulatory scope of consolidation as set out in paragraph 8, the amount of any investment in the capital of that entity that is totally or partially deducted from CET1 capital or from Additional Tier 1 capital of the bank following the corresponding deduction approach in paragraphs 84 to 89 of the Basel III framework may also be deducted from the exposure measure. For banks using the internal ratings-based (IRB) approach to determining capital requirements for credit risk, paragraph 73 of the Basel III framework requires any shortfall in the stock of eligible provisions relative to expected losses to be deducted from CET1 capital. The same amount may be deducted from the exposure measure.

According to the above, Table 6.15 summarizes the components of the leverage ratio and its evolution during the stress testing exercise. The minimum 3% is exceeded during the entire stress testing period. In the following sections liquidity is examined by our focusing on the LCR and the NSFR.

TABLE 6.15 Bank Alpha’s Leverage Ratio ($ Billions) t0

t1

t2

t3

On-balance sheet items

100.00

105.30

109.62

114.18

Assets deducted from tier 1 capital

−1.56

−1.61

−1.66

−1.71

Total balance sheet exposure

98.44

103.69

107.96

112.47

Tier 1 capital

5.44

4.57

4.79

5.44

Leverage ratio (%)

5.53

4.41

4.44

4.84

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 219

6.4.3 Liquidity Coverage Ratio The LCR identifies the amount of high-quality liquid assets that an institution holds to offset the net cash outflows (operational liquidity risk). The LCR assumes an acute short-term, 30-day stress scenario. The specified scenario entails both bank-specific and market-wide shocks built on actual circumstances experienced during the financial crisis. The quantification of liquid asset requirements and the formulation of qualitative needs is undertaken by consideration of observed trends in the value of assets under stressed conditions. Additionally, the expected and observed behaviors of inflows and outflows during periods of unexpected volatility and crisis need to be taken into account. One needs to consider their capacity to generate cash through sale or secured borrowing to test the quality of liquid assets . The test is favorable when no loss of value is experienced in periods of severe idiosyncratic and market stress. High-quality assets should ideally be eligible at central banks for intraday liquidity needs and overnight facilities. The LCR is built in terms of the following proportion: LCR =

High-quality liquid assets Total net cash outflow

,

(6.7)

where the numerator and denominator of Eq. (6.7) are obtained as detailed below: ●

High-quality liquid assets. This pool comprises readily marketable securities with the potential to generate liquidity under stress. The buffer of highquality liquid assets comprises level 1 and level 2 assets as detailed below. Level 1 instruments can be included without limit, while a 40% limit (out of the total) is applied to level 2 assets. This nonexhaustive list aims to pinpoint the key items to be considered. – Level 1 assets. These instruments are included in the buffer at their market value. The following assets are assigned a factor of 100% reflecting a presumed high liquidity under stress: * Cash. * Central bank reserves able to be drawn down in times of stress. * Liquid, marketable securities issued by or guaranteed by sovereign states, central banks, and certain international organizations and that qualify for a 0% risk weight under the Basel II standardized approach for credit risk. * Certain nonzero RWAs may also be included where these match an institution’s jurisdictional currency liquidity needs or operational requirements. – Level 2 assets. These assets can be included subject to a minimum 15% supervisory haircut to their market value. More specifically, a distinction

220 Stress Testing and Risk Integration in Banks

is made between level 2A and level 2B assets, having haircuts of 15% and 50% respectively. In all cases, level 2 assets are capped at 40% (after haircut) of the total buffer. The following list focuses on level 2A items (certain additional assets may be included in level 2 at the discretion of national authorities): * Liquid, marketable securities issued by or guaranteed by sovereign states, central banks, and certain international organizations and that qualify for a 20% risk weight under the Basel II standardized approach for credit risk. * Certain corporate bonds (senior status, vanilla) of at least an AA rating or equivalent. * Covered bonds of at least an AA rating or equivalent. ●

Total net cash outflow. The denominator of the LCR relies on cumulative outflows deducted from inflows over a 30-day stress period. Inflows are capped at 75% of outflows. A weighting scheme is designed to assign a specific weight (run-off rate) to each outflow and inflow category. In what follows, a nonexhaustive list is shown to highlight the main items to be considered: – Outflows: * For stable retail deposits, a factor of 5% (run-off rate) is applied. * For less stable retail deposits, the run-off rate is 10%. * Unsecured wholesale funding with operational relationships is subject to a 25% run-off rate. – Inflows: * Maturing reserve repo or securities borrowing transactions secured by level 1 assets are subject to a 0% cash inflow rate. Lines of credit, liquidity facilities, and other contingent funding receive a 0% inflow rate. * For level 2 collateral, the rate is 15%. * For retail and small business inflows, the rate is 50%. * For nonlevel 1 or nonlevel 2 assets, the ratio is 100%.

The next section focuses on Bank Alpha’s LCR during the stress testing exercise.

6.4.4 Bank Alpha’s Stress Testing Liquidity Coverage Ratio Bank Alpha is investigated from the regulatory liquidity perspective as detailed in Table 6.16. Assets and liabilities are assigned a factor according to what was described in Section 6.4.3. More specifically, a factor of 100% is applied to cash and trading account instruments (the latter have a 0% risk weight under the Basel II standardized approach). A 15% haircut (i.e., factor of 85%) is applied to interestbearing deposits toward high-rating banks. On the total net cash outflow side, a

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 221

TABLE 6.16 Bank Alpha’s Liquidity Coverage Ratio ($ Billions) Factor (%) t0

A1 assets

A2 assets

2.00

100.00

2.00

Trading account

3.00

100.00

3.00

Interest-bearing deposits

3.00

85.00

2.55

Retail

45.00

5.00

2.25

Wholesale

25.00

25.00

6.25

Total liquid assets Outflows

7.55

Total outflows Inflows

Total inflows

8.50 2.50

50.00

Net outflows

104.14

A1 assets

Cash

A2 assets

2.00

100.00

Trading account

3.00

100.00

3.00

Interest-bearing deposits

3.90

85.00

3.32

Total liquid assets Outflows

Retail

50.03

5.00

2.50

Wholesale

25.00

25.00

6.25

Total inflows

8.75 0.97

50.00

Net outflows

A1 assets

A2 assets

100.60 Cash

2.00

100.00

2.00

Trading account

3.00

100.00

3.00

Interest-bearing deposits

3.90

85.00

3.32

Retail

53.31

5.00

2.67

Wholesale

25.00

25.00

6.25

Total liquid assets Outflows

8.32

Total outflows Inflows Net outflows LCR (%)

0.48 8.27

LCR (%) t2

2.00

8.32

Total outflows Inflows

1.25 7.25

LCR (%) t1

Assets/Flow

Cash

Total inflows

8.92 1.66

50.00

0.84 8.08 102.97 (Continued)

222 Stress Testing and Risk Integration in Banks

TABLE 6.16 Bank Alpha’s Liquidity Coverage Ratio ($ Billions)—cont’d Factor (%) t3

A1 assets

A2 assets

2.00

100.00

2.00

Trading account

3.00

100.00

3.00

Interest-bearing deposits

3.90

85.00

3.22

Retail

56.40

5.00

2.82

Wholesale

25.00

25.00

6.25

Total liquid assets Outflows

8.32

Total outflows Inflows

Assets/Flow

Cash

Total inflows

Net outflows LCR (%)

9.07 2.16

50.00

1.08 7.99 104.13

LCR, Liquidity coverage ratio.

distinction is made between outflows and inflows. Retail outflows are assigned a factor of 5% because of their stable nature. In contrast, a factor of 25% is applied to wholesale (less stable) deposits. From the inflow standpoint, a factor of 50% is applied because of the retail and small business counterpart nature of the flows. Additionally, the 75% outflow cap is never triggered by inflows. Bank Alpha highlights a solid liquidity structure mainly due to the nature of its assets and liabilities. The LCR exceeds the 100% threshold during all of the stress testing exercise. Nonetheless, Chapter 7 will show that under some conditions deposit funding instability may undermine bank resilience. Once completing the LCR analysis, the next section focuses on the net stable funding ratio.

6.4.5 Net Stable Funding Ratio The NSFR aims to ensure that banks hold a minimum amount of stable funding to run their business. It is based on the liquidity characteristics of their cash outflows and inflows over a 1-year horizon. This ratio has been introduced to reduce maturity mismatches between assets and liabilities, to cover an extended firm-specific stress scenario. In such a context, a bank may encounter a significant decline in its profitability or solvency arising from its risk profile. Additionally, a potential downgrade of its debt or deposit rating may call into question the credit quality of the institution. Under this kind of scenario, borrowings become difficult and usual open market operations may be subject to haircuts.

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 223

The NSFR is computed as follows: NSFR =

Available stable funding , Required stable funding

(6.8)

where items classifiable among available stable funding (ASF) and required stable funding (RSF) are listed below: ●

ASF. Stable funding is defined as the portion of those types of equity and liability financing expected to provide reliable sources of funds over a 1-year time horizon to cover conditions of extended stress. Firstly, the carrying value of bank equity and liabilities is assigned to one of the following five categories: – Total regulatory capital (excluding tier 2 instruments with residual maturity of less than 1 year); other capital instruments and liabilities with effective residual maturity of 1 year or more. The ASF factor is 100%. – Stable nonmaturity (demand) deposits and term deposits with residual maturity of less than 1 year provided by retail and small business customers. The ASF factor is 95%. – Less stable nonmaturity deposits and term deposits with residual maturity of less than 1 year provided by retail and small business customers. The ASF factor is 90%. – Funding with residual maturity of less than 1 year provided by nonfinancial corporate customers; operational deposits and funding with residual maturity of less than 1 year from sovereign states, public sector entities, and multilateral and national development banks; other funding with residual maturity between 6 months and less than 1 year not included in the above categories, including funding provided by central banks and financial institutions. The ASF factor is 50%. – All other liabilities and equity categories not included in the above categories have an ASF factor of 0%. Then the amount assigned to each category needs to be multiplied by an ASF factor ranging from 0% to 100%. The total ASF is the sum of the weighted amounts.

●

RSF. The amount of stable funding required by supervisors is measured by taking into account the characteristics of the liquidity risk profiles of assets and off-balance sheet exposures. The RSF is calculated as the sum of the value of assets held, multiplied by a specific factor assigned to each asset type. Some of the major categories are listed below: – Coins and banknotes, central bank reserves, and claims on central banks with residual maturities of less than 6 months. The RSF factor is 0%. – Unencumbered level 1 assets, excluding coins, banknotes, and central bank reserves. The RSF factor is 5%. – Unencumbered loans to financial institutions with residual maturities of less than 6 months, where the loan is secured against level 1 assets

224 Stress Testing and Risk Integration in Banks

–

–

–

–

–

and where the bank has the ability to freely prehypothecate the received collateral for the life of the loan. The RSF factor is 10%. All other unencumbered loans to financial institutions with residual maturities of less than 6 months not included in the above categories (unencumbered level 2A assets). The RSF factor is 15%. Unencumbered level 2B assets; high-quality liquid assets encumbered for a period of 6 months or more and less than 1 year; loans to financial institutions and central banks with residual maturities between 6 months and less than 1 year; deposits held at other financial institutions for operational purposes; all other assets not included in the above categories with residual maturity of less than 1 year, including loans to nonfinancial corporate clients, loans to retail and small business customers, and loans to sovereign states and public sector entities. The RSF factor is 50%. Unencumbered residential mortgages with a residual maturity of 1 year or more and with a risk weight of less than or 35% under the standardized approach to credit risk; other unencumbered loans not included in the above categories, excluding loans to financial institutions, with a residual maturity of 1 year or more and with a risk weight of less than or 35% under the standardized approach. The RSF factor is 65%. Cash, securities, or other assets posted as the initial margin for derivative contracts and cash or other assets provided as contributions to the default fund of a clearing house; other unencumbered performing loans with risk weights greater than 35% under the standardized approach and residual maturities of 1 year or more, excluding loans to financial institutions; unencumbered securities that are not in default and do not qualify as high-quality liquid assets with a remaining maturity of 1 year or more and exchange-traded equities; physical traded commodities, including gold. The RSF factor is 85%. All other assets not included in the above categories, including nonperforming loans, loans to financial institutions with a residual maturity of 1 year or more, nonexchange-traded equities, and fixed assets; items deducted from regulatory capital, retained interest, insurance assets, subsidiary interests, and defaulted securities. The RSF factor is 100%.

The next section focuses on Bank Alpha’s NSFR during the stress testing exercise.

6.4.6 Bank Alpha’s Stress Testing Net Stable Funding Ratio Bank Alpha is investigated from the NSFR perspective as detailed in Table 6.17. Assets and liabilities are assigned a factor according to the details provided in Section 6.4.5. In more detail, a distinction is made between assets and liabilities. On the asset side, a factor of approximately 21% characterizes cash resources during the stress testing exercise. This is due to a mix of cash, central bank reserves, and loans to financial institutions with residual maturity between 6

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 225

TABLE 6.17 Bank Alpha’s Stress Testing Net Stable funding Ratio Dynamics ($ Billions) Exposure t0

Cash resources

Average Factor (%)

8.00

20.63

1.65

Securities

14.00

100.00

14.00

Loans

70.00

77.50

54.25

8.00

100.00

Other assets Total RSF Deposits

70.00

74.64

55.25

Other liabilities

17.00

26.47

4.50

Subordinated debts

4.00

75.00

3.00

Noncontrolling interests

2.00

100.00

2.00

Shareholder equity

7.00

100.00

7.00

Total ASF

68.75

NSFR (%) t1

Cash resources

88.25 9.00

20.50

1.85

Securities

15.00

100.00

15.00

Loans

72.28

77.20

55.80

9.02

100.00

Other assets Total RSF

t2

8.00 77.90

9.02 81.67

Deposits

75.04

79.00

59.28

Other liabilities

18.08

26.05

4.71

Subordinated debts

4.00

100.00

4.00

Noncontrolling interests

2.00

100.00

2.00

Shareholder equity

6.18

100.00

6.18

Total ASF

76.17

NSFR (%)

93.27

Cash resources

9.00

21.00

1.89

Securities

15.00

100.00

15.00

Loans

75.43

77.20

58.23

Other assets

10.19

100.00

10.19

Total RSF

85.31

Deposits

78.31

77.86

60.97

Other liabilities

18.86

27.50

5.19

4.00

100.00

Subordinated debts

4.00 (Continued)

226 Stress Testing and Risk Integration in Banks

TABLE 6.17 Bank Alpha’s Stress Testing Net Stable funding Ratio Dynamics ($ Billions)—cont’d Exposure

t3

Average Factor (%)

Noncontrolling interests

2.00

100.00

2.00

Shareholder equity

6.45

100.00

6.45

Total ASF

78.61

NSFR (%)

92.15

Cash resources

9.00

21.00

1.89

Securities

15.00

100.00

15.00

Loans

78.52

78.05

61.28

Other assets

11.66

100.00

11.66

Total RSF

89.83

Deposits

81.40

77.07

62.73

Other liabilities

19.63

27.40

5.38

Subordinated debts

4.00

100.00

4.00

Noncontrolling interests

2.00

100.00

2.00

Shareholder equity

7.15

100.00

7.15

Total ASF

81.26

NSFR (%)

90.46

ASF, Available stable funding; NSFR, net stable funding ratio; RSF, required stable funding.

months and 1 year. Securities and other assets not classifiable among other most favorable exposures are assigned a factor of 100%. Finally, the wide range of loans included in Bank Alpha’s balance sheet have a weight oscillating around 78% during the 3-year stress testing period. On the liability side, a high quota of deposits is classified among stable nonmaturity and less stable nonmaturity deposits. Additionally, funding with residual maturity of less than 1 year provided by nonfinancial customers causes the ASF factor to fluctuate around 78% along the stress testing horizon. A mix of different facilities characterizes other liabilities, to which a factor oscillating around 27% is assigned. $1 billion subordinated debts have a factor of 0% at t0 due to their expiry during the period [t0 , t1 ]. In contrast, a 100% weight is applied to all others and from t1 onward because of their long term maturity. All other equity instruments are assigned a 100% weight. All in all, Table 6.17 highlights a substantial balance between assets and liabilities. It is also coherent with the asset and liability management liquidity

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 227

analysis described in Chapter 3. Nonetheless, the assumption of stable deposits is critical for the overall assessment. This will be illustrated in Chapter 7 when we focus on a fully integrated liquidity analysis. Example 6.3 helps us understand liquidity strategic planning based on the LCR and NSFR. Example 6.3 LCR and NSFR The chief executive officer of a bank asks the chief risk officer to define strategies to improve the LCR and NSFR to reach a 100% target. The starting point is the balance sheet shown in Table 6.18. Table 6.19 shows the weights to be applied to each asset and liability category to compute both the LCR and the NSFR. The 75% outflow limit is applied to inflows. Moreover, all inflows can be deducted from outflows. According to the weighting scheme described in Section 6.4.3, weighted liquid assets amount to $13 million, while net outflows are $59 million (i.e., $62 billion outflows – $3 billion inflows). The LCR is 13.00 59.00 = 22.03%. With respect to the NSFR, available stable funds amount to $66.75 million, while requested stable funds amount to $75.75 million. The NSFR is 66.75 75.75 = 88.12%. The first strategy considered by the chief risk officer is to deleverage $20 million by selling corporate credit exposures and correspondingly reducing other funding having maturity within the year. This idea is outlined in Table 6.20. According to Table 6.20, this strategy would allow the bank to improve its NSFR as follows: 56.75 55.75 = 102%. However the LCR would not be substantially affected: 13.00 = 33%. Hence additional changes are required to achieve an LCR of 100% 39.00 or greater. The chief risk officer considers an additional increase of stable deposits to finance high-quality, unencumbered liquid assets (e.g., bank deposits) as detailed in Table 6.21 During the meeting with the chief executive officer, the chief risk officer presents the following two-step proposal: ● Deleverage by selling $30 million of credits and correspondingly reduce shortterm funds (the original strategy was based on $20 million). ● Stable deposits increase by $17 million to finance bank deposits. 67.90 This strategy would allow the bank to improve its NSFR ( 45.75 = 148%) and hit the LCR target ( 30.00 = 101%). 29.85

6.5 SUMMARY An introduction to the definition of own funds paved the way to understanding how to compute capital ratios. The distinction between tier 1 common equity, additional tier 1 capital, and tier 2 capital was described by our highlighting the key differences between accounting and regulatory capital. As part of the denominator of capital ratios, market, credit, and operational RWAs were examined from a silo perspective. Then they were aggregated to compute the overall

228 Stress Testing and Risk Integration in Banks

TABLE 6.18 Bank’s Balance Sheet ($ Millions) Assets Cash resources

8.00 Cash

5.00

Central bank reserves

3.00

Securities

17.00 Securities (0% RWAs)

10.00

Corporate bonds >1 year

7.00

Corporate credits >1 year

30.00

Loans

75.00

Loan portfolio corporate <1 year

5.00

Loan portfolio retail < 1year

25.00

Mortgage loans (35% RWAs)

15.00

Total assets

100.00

Liabilities Deposits

25.00 Stable deposits Less stable deposits

20.00 5.00

Other liabilities

Subordinated debts

55.00 Corporate operational

10.00

Bonds > 1 year

10.00

Other funding < 1 year

35.00 10.00

Noncontrolling interests

5.00

Shareholder equity

5.00

Total liabilities

100.00

RWAs, Risk-weighted assets.

bank RWAs by our considering regulatory constraints (e.g., floor, shortfall, and so on). The illustrative example of Bank Alpha was useful to investigate how regulatory rules impact the RWAs. A specific focus targeted the credit risk assessment by comparing the expected loss, provisions, and best estimate of the expected loss. The computation of capital ratios finalized the stress testing process illustrated throughout the book. Then, leverage and liquidity ratios highlighted the importance of monitoring additional nonrisk-based measures to

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 229

TABLE 6.19 Liquidity Coverage Ratio and Net Stable Funding Ratio Analysis ($ Millions) Assets Balance Cash resources

LCR Assets

Inflows

RSFs

8.00 Cash

5.00

100.00%

0.00%

Central bank reserves

3.00

100.00%

0.00%

Securities (0% RWAs)

5.00

100.00%

5.00%

12.00

0.00%

100.00%

Securities

17.00

Corporate bonds > 1 year Loans

75.00 Corporate credits > 1 year

30.00

0.00%

100.00%

Loan portfolio corporate < 1 year

5.00

1.00

50.00%

Loan portfolio retail < 1 year

25.00

2.00

85.00%

Mortgage loans (35% RWAs)

15.00

Total Assets

0.00%

100.00

13.00

65.00%

3.00

75.75

Liabilities Balance Deposits

LCR outflows

ASFs

10.00 Stable deposits

5.00

5.00%

95.00%

Less stable deposits

5.00

10.00%

90.00%

25.00

25.00%

50.00%

5.00

100.00%

100.00%

40.00

100.00%

50.00%

10.00

100.00%

100.00%

Noncontrolling interests

5.00

0.00%

100.00%

Shareholder equity

5.00

0.00%

100.00%

Other liabilities

70.00 Corporate operational Bonds >1 year Other funding <1 year

Subordinated debts

Total liabilities

100.00

62.00

66.75

ASFs, Available stable funds; LCR, liquidity coverage ratio; RSFs, required stable funds; RWAs, risk-weighted assets.

230 Stress Testing and Risk Integration in Banks

TABLE 6.20 Deleveraging Strategy ($ Millions) Assets Balance Cash resources

LCR Assets

Inflows RSFs

8.00 Cash

5.00

100.00%

0.00%

Central bank reserves

3.00

100.00%

0.00%

Securities (0% RWAs)

5.00

100.00%

5.00%

Corporate bonds > 1 year 12.00

0.00%

100.00%

0.00%

100.00%

Securities

17.00

Loans

55.00 Corporate credits > 1 year Loan portfolio corporate < 1 year

10.00 5.00

1.00

50.00%

Loan portfolio retail < 1 25.00 year

2.00

85.00%

Mortgage loans (35% RWAs)

15.00

Total assets

0.00% 80.00

13.00

65.00% 3.00

55.75

Liabilities Balance Deposits

ASFs

10.00 Stable deposits

5.00

5.00%

95.00%

Less stable deposits

5.00

10.00%

90.00%

25.00%

50.00%

5.00

100.00%

100.00%

20.00

100.00%

50.00%

10.00 100.00%

100.00%

Other liabilities

50.00 Corporate operational Bonds >1 year Other funding <1 year

Subordinated debts

LCR Outflows

25.00

Noncontrolling interests

5.00

0.00%

100.00%

Shareholder equity

5.00

0.00%

100.00%

Total liabilities

80.00

42.00

ASFs, Available stable funds; LCR, liquidity coverage ratio; RSFs, required stable funds; RWAs, risk-weighted assets.

56.75

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 231

TABLE 6.21 Deleveraging and Deposit Increase Strategy ($ Millions) Assets Balance Cash resources

LCR Assets

Inflows RSFs

25.00 Cash

22.00

100.00%

0.00%

Central bank reserves

3.00

100.00%

0.00%

Securities (0% RWAs)

5.00

100.00%

5.00%

12.00

0.00%

100.00%

Securities

17.00

Corporate bonds >1 year Loans

45.00 Corporate credits >1 year –

0.00%

100.00%

Loan portfolio corporate <1 year

5.00

1.00

50.00%

Loan portfolio retail <1 year

25.00

2.00

85.00%

Mortgage loans (35% RWAs)

15.00

Total assets

0.00% 87.00

30.00

65.00% 3.00

45.75

Liabilities Balance Deposits

ASFs

27.00 Stable deposits Less stable deposits

22.00

5.00%

95.00%

5.00

10.00%

90.00%

25.00

25.00%

50.00%

5.00

100.00%

100.00%

10.00

100.00%

50.00%

10.00 100.00%

100.00%

Other liabilities

40.00 Corporate operational Bonds >1 year Other funding <1 year

Subordinated debts

LCR Outflows

Noncontrolling interests

5.00

0.00%

100.00%

Shareholder equity

5.00

0.00%

100.00%

Total liabilities

87.00

32.85

67.90

ASFs, Available stable funds; LCR, liquidity coverage ratio; RSFs, required stable funds; RWAs, risk-weighted assets.

232 Stress Testing and Risk Integration in Banks

ensure a bank is solvent both in the long run and in the short run. A structural balance between own resources and investment together with equilibrated cash flow mismatches finalized the regulatory stress testing analysis.

SUGGESTIONS FOR FURTHER READING A regulatory perspective inspired this chapter. Therefore the key references relate to the Basel Accord and its methodological notes. The reader may find it useful to read BIS (2011, 2014a,b) and the most recent consultation papers where potential changes to the current regulatory framework are under discussion (BIS, 2015).

EXERCISES Exercise 6.1 Let us consider the balance sheet shown in Table 6.22. A capital planning exercise is required aimed at improving the core capital ratio by 1.5% and the total capital ratio by 2.0% in the next 2 years. The bank relies on advanced methods to compute RWAs for market and credit risk, while for the operational risk a basic approach is used. The following simplistic assumptions characterize the computation. ●

Market risk. The value at risk may be computed by use of the following synthetic measures of volatility: shares have a 0.05% daily variance, while for bonds it is 0.01%. Their covariance is 0.02%.

TABLE 6.22 Balance Sheet at t0 ($ Billions) Assets

Liabilities

Cash resources

2.00 Deposits

60.00

Securities

8.00 Bonds

18.00

Shares

3.00

Subordinated debts

15.00

Bonds

5.00

Shareholder equity

7.00

Loans

90.00 SME

35.00

Corporate 55.00 Total assets

Common shares

3.00

Preferred shares

1.00

Retained earnings 3.00 100.00 Total liabilities

SME, Small and medium-sized enterprise.

100.00

Regulatory Capital, RWA, Leverage, and Liquidity Requirements Chapter | 6 233

●

●

Credit risk. For small- and medium-sized enterprises the average probability of default is 4%, while the loss given default is 25%. In contrast, the average corporate probability of default is 2% and the loss given default is 20%. Operational risk. The gross income for the previous 3 years expressed in billion dollars is 0.80, 0.70, and −0.10.

The reader is invited to adopt other suitable assumptions required to perform the capital planning exercise. Exercise 6.2 Let us consider the balance sheet structure of the bank outlined in Exercise 6.1. A coherent liquidity planning is required to achieve the 100% minimum LCR and NSFR based on the t0 factors as detailed in Table 6.23. Solutions are available at www.tizianobellini.com.

TABLE 6.23 Liquidity Coverage Ratio and Net Stable Funding Ratio Factors at t0 Assets LCR Assets (%) Cash resources

Inflows

100.00

RSFs (%) 0.00

Securities Shares

0.00

100.00

Bonds

100.00

100.00

Loans SME

5.00

100.00

Corporate

2.00

100.00

Liabilities LCR Outflows (%)

ASFs (%)

Deposits

10.00

90.00

Bonds

100.00

100.00

Subordinated debts

100.00

100.00

Common shares

0.00

100.00

Preferred shares

0.00

100.00

Retained earnings 0.00

100.00

Shareholder equity

ASFs, Available stable funds; LCR, liquidity coverage ratio; RSFs, required stable funds; SME, small- and medium-sized enterprise.

234 Stress Testing and Risk Integration in Banks

REFERENCES BIS, 1988. International Convergence of Capital Measurement and Capital Standards. Bank for International Settlements, Basel. BIS, 2004. Modifications to the capital treatment for expected and unexpected credit losses in the new Basel accord. Bank for International Settlements, Basel. BIS, 2006. Basel II International Convergence of Capital Measurement and Capital Standards: A Revised Framework, Bank for International Settlements, Basel. BIS, 2011. Basel III: A global regulatory framework for more resilient banks and banking systems. Bank for International Settlements, Basel. BIS, 2014a. Regulatory Consistency Assessment Programme (RCAP), Assessment of Basel III regulations—United States of America. Bank for International Settlements, Basel. BIS, 2014b. Revised Basel III leverage ratio framework and disclosure requirements. Bank for International Settlements, Basel. BIS, 2015. Consultative document: Capital floors: the design of a framework based on standardised approaches. Bank for International Settlements, Basel. BOE, 2013. The Basel I Floor Supervisory Statement, SS8/13, London. FRB, 2015. Comprehensive Capital Analysis and Review 2015: assessment framework and results Board of Governors of the Federal Reserve System, Washington, DC.